From: Jeff King <peff@peff.net>
Date: Thu, 11 Apr 2013 05:31:45 GMT
Subject: Re: [PATCH 2/5] check-ignore: allow incremental streaming of queries via --stdin
Message-ID: <20130411053145.GB28915@sigill.intra.peff.net>
In-Reply-To: <1365645575-11428-2-git-send-email-git@adamspiers.org>

On Thu, Apr 11, 2013 at 02:59:32AM +0100, Adam Spiers wrote:

> @@ -111,14 +110,11 @@ static int check_ignore_stdin_paths(struct path_exclude_check check, const char
>  				die("line is badly quoted");
>  			strbuf_swap(&buf, &nbuf);
>  		}
> -		ALLOC_GROW(pathspec, nr + 1, alloc);
> -		pathspec[nr] = xcalloc(strlen(buf.buf) + 1, sizeof(*buf.buf));
> -		strcpy(pathspec[nr++], buf.buf);
> +		pathspec[0] = xcalloc(strlen(buf.buf) + 1, sizeof(*buf.buf));
> +		strcpy(pathspec[0], buf.buf);
> +		num_ignored += check_ignore(check, prefix, (const char **)pathspec);
> +		maybe_flush_or_die(stdout, "check-ignore to stdout");

Now that you are not storing the whole pathspec at once, the pathspec
buffer only needs to be valid for the length of check_ignore, right?
That means you can drop this extra copy and just pass in buf.buf:

  pathspec[0] = buf.buf;
  num_ignored += check_ignore(check, prefix, pathspec);

> +test_expect_success 'setup: have stdbuf?' '
> +	if which stdbuf >/dev/null 2>&1
> +	then
> +		test_set_prereq STDBUF
> +	fi
> +'

Hmm. Today I learned about stdbuf. :)

-Peff